- keep failed overlong coordinator outputs in the conversation, suppress compaction notices, and surface a retry action in the Auto Drive card so follow-up requests retain the error context
- add explicit Action app event, pass compaction notice intent through to the TUI, and raise the coordinator worker stack to 1 MiB with an opt-in backtrace hook for debugging validation overflows
- simplify bottom_spacer_hysteresis_requests_followup_frame to a known viewport pair to avoid FD exhaustion during repeated harness setup
Fixes#432Fixes#430Fixes#433Fixes#429
Replace the retry Thinking event text with a neutral message so the Auto Drive card shows the retry without exposing prompt-length guidance to the user. Build fast remains green.
fixes#432
Add assistant retry note into the conversation so the Auto Drive card shows context while retrying overlong prompt_sent_to_cli errors. Developer guidance string remains in the conversation; raw model output stays in assistant message.
fixes#432
Assistant message now keeps the raw model output; developer note carries the fixed guidance string so the coordinator sees both the failure context and the explicit <=600 notice. Build fast remains green.
fixes#432
Replace the retry assistant insert with a fixed error string so the CLI clearly sees the overlong prompt error without echoing the long prompt. Persist the message in the conversation and simplify developer notes.
fixes#432
Surface the invalid assistant JSON as an assistant message before retrying recoverable validation errors (e.g., overlong prompt_sent_to_cli) so the coordinator can see what failed. Retains existing retry guidance path.
Refs #432
* emit json snapshots once then deltas with ordering metadata
* dedupe browser snapshots and update history pruning
* refresh mac temp path tests and TUI snapshots for new ordering
build-fast.sh
cargo nextest run --no-fail-fast --manifest-path code-rs/Cargo.toml
James Peter